1. Technical Field
The present invention relates to a terminal apparatus and a retransmission control method.
2. Description of the Related Art
3GPP long term evolution (LTE) adopts orthogonal frequency division multiple access (OFDMA) as a downlink communication scheme. In a radio communication system to which 3GPP LTE is applied, a base station transmits a synchronization signal (synchronization channel: SCH) and a broadcast signal (broadcast channel: BCH) using predetermined communication resources. A terminal first secures synchronization with the base station by catching an SCH. Then, the terminal acquires parameters (e.g. frequency bandwidth) specific to the base station by reading BCH information (see Non-Patent Literatures 1, 2, and 3).
Furthermore, after completing the acquisition of the parameters specific to the base station, the terminal transmits a connection request to the base station and establishes communication with the base station. The base station transmits control information to the terminal with which communication is established through a physical downlink control channel (PDCCH) as necessary.
The terminal then makes a “blind decision” on each of a plurality of pieces of control information included in the received PDCCH signal. That is, the control information includes a cyclic redundancy check (CRC) portion, and this CRC portion is masked with a terminal ID of a transmission target terminal in the base station. Therefore, the terminal is difficult to decide whether or not the control information is directed to its own terminal until the CRC portion of the received control information is demasked with the terminal ID of the terminal. In the blind decision, when a demasking result represents that a CRC calculation is OK, it is determined that the control information is directed to its own terminal.
Furthermore, in 3GPP LTE, automatic repeat request (ARQ) is applied to downlink data from a base station to a terminal. That is, the terminal feeds back a response signal indicating an error detection result of downlink data to the base station. The terminal performs a CRC on the downlink data, and feeds back acknowledgment (ACK) when CRC=OK (no error) and negative acknowledgment (NACK) when CRC=NG (error) to the base station as a response signal. A binary phase shift keying (BPSK) scheme is used for modulation of the response signal (that is, the ACK/NACK signal). Further, an uplink control channel such as a physical uplink control channel (PUCCH) is used for feedback of the response signal. When the received response signal represents NACK, the base station transmits retransmission data to the terminal.
Here, the control information transmitted from the base station includes resource assignment information including resource information and the like assigned from the base station to the terminal. The PDCCH is used for transmission of this control information as described above. The PDCCH is configured with one or more L1/L2 control channels (L1/L2 CCHs). Each L1/L2 CCH is configured with one or more control channel elements (CCEs). That is, a CCE is a base unit for mapping control information to a PDCCH. Furthermore, when one L1/L2 CCH is configured with a plurality of CCEs, a plurality of CCEs whose indices are consecutive are assigned to the L1/L2 CCH. The base station assigns an L1/L2 CCH to a resource assignment target terminal according to the number of CCEs necessary for notifying control information to the resource assignment target terminal. The base station then transmits the control information mapped to a physical resource corresponding to the CCE of the L1/L2 CCH.
Here, each CCE has a one-to-one correspondence with a component resource of the PUCCH. Therefore, the terminal that has received the L1/L2 CCH can implicitly specify a component resource of the PUCCH corresponding to the CCEs configuring the L1/L2 CCH, and transmits a response signal to the base station using the specified resource. This allows downlink communication resources to be used efficiently.
As illustrated in FIG. 1, a plurality of response signals transmitted from a plurality of terminals are spread by a Zero Auto-correlation (ZAC) sequence having a Zero Auto-correlation characteristic, a Walsh sequence, and a discrete Fourier transform (DFT) sequence on a time axis, and code-multiplexed within the PUCCH. In FIG. 1, (W0, W1, W2, W3) represents a Walsh sequence (which may be also referred to as “Walsh code sequence” or “Walsh code”) having a sequence length of 4, and (F0, F1, F2) represents a DFT sequence having a sequence length of 3. As illustrated in FIG. 1, in the terminal, a response signal of ACK or NACK is first primary-spread to frequency components corresponding to a one single carrier frequency division multiple access (1 SC-FDMA) symbol on a frequency axis by a ZAC sequence (having a sequence length of 12). Next, the response signal subjected to the primary spreading and the ZAC sequence functioning as a reference signal are secondary-spread in association with a Walsh sequence (having a sequence length 4: W0 to W3) and a DFT sequence (having a sequence length 3: F0 to F2) respectively. Further, the signal subjected to the second spreading is transformed into a signal having a sequence length of 12 on the time axis by the inverse fast Fourier transform (IFFT). Then, a cyclic prefix (CP) is added to the signal that has been subjected to the IFFT, and thus a one-slot signal including 7 SC-FDMA symbols is generated.
Here, response signals transmitted from different terminals are spread using sequences corresponding to different cyclic shift indices or orthogonal cover (OC) indices (that is, a set of a Walsh sequence and a DFT sequence). Therefore, the base station can demultiplex a plurality of code-multiplexed response signals using a conventional dispreading process and a conventional correlation process (see Non-Patent Literature 4).
However, since each terminal makes a blind decision on a downlink assignment control signal in each subframe directed to its own terminal, the terminal side does not necessarily succeed in receiving the downlink assignment control signal. When the terminal fails to receive the downlink assignment control signal directed to its own terminal in a certain downlink unit band, the terminal is difficult to know whether or not there is downlink data, directed to its own terminal, in the downlink unit band. Therefore, when failing to receive the downlink assignment control signal in a certain downlink unit band, the terminal is difficult to generate a response signal on the downlink data in the downlink unit band. This error case is defined as discontinuous transmission (DTX) of a response signal (DTX of ACK/NACK signals) in the sense that the terminal side does not transmit the response signal.
Meanwhile, the uplink control channel (PUCCH) is also used for transmission of a scheduling request (SR) (which may be also represented by a scheduling request indicator (SRI)) which is an uplink control signal indicating that uplink data to be transmitted from the terminal side has been generated. When a connection with the terminal has been established, the base station individually assigns a resource to be used for transmission of the SR (hereinafter, referred to as “SR resource”) to each terminal. Further, an on-off keying (OOK) scheme is applied to the SR, and the base station detects the SR from the terminal based on whether or not the terminal is transmitting an arbitrary signal using the SR resource. Further, the SR is spread using a ZAC sequence, a Walsh sequence, and a DFT sequence in the same manner as the above-mentioned response signal.
In the LTE system, the SR and the response signal may be generated in the same sub frame. In this case, when the terminal code-multiplexes and transmits the SR and the response signal, a peak to average power ratio (PAPR) of a synthesized waveform of a signal transmitted from the terminal significantly deteriorates. However, in the LTE system, since importance is put on amplification efficiency of the terminal, when the SR and the response signal have been generated in the same sub frame at the terminal side, the terminal transmits the response signal (response signals illustrated in FIGS. 2A to 2D) using the SR resource previously individually assigned to each terminal, without using a resource (hereinafter, referred to as “ACK/NACK resource”) used for transmission of the response signal as illustrated in FIG. 2A.
That is, when the terminal side has only to transmit only a response signal (“when only response signal is transmitted” illustrated in FIG. 2C), the terminal transmits the response signal (a response signal illustrated in FIG. 2C) using the ACK/NACK resource. On the other hand, when the SR and the response signal have been generated in the same sub frame at the terminal side (“when response signal and SR are transmitted” illustrated in FIG. 2D), the terminal transmits the response signal (a response signal illustrated in FIG. 2D) using the SR resource.
Thus, the PAPR of the synthesized waveform of the signal transmitted from the terminal can be reduced. At this time, the base station detects the SR from the terminal based on whether or not the SR resource is being used. In addition, the base station determines whether or not the terminal has transmitted either ACK or NACK, based on a phase (that is, a BPSK demodulation result) of a signal transmitted through the SR resource (the ACK/NACK resource when the SR resource is not used).
Further, the standardization of 3GPP LTE-advanced that realizes faster communication than 3GPP LTE has started. A 3GPP LTE-advanced system (which may also be hereinafter referred to as “LTE-A system”) follows the 3GPP LTE system (which may also be hereinafter referred to as “LTE system”). In order to realize a downlink transmission rate of a maximum of 1 Gbps or above, 3GPP LTE-advanced is expected to introduce base stations and terminals capable of performing communication at a wideband frequency of 40 MHz or above.
In an LTE-A system, in order to simultaneously realize communication at an ultra-high transmission rate several times as fast as a transmission rate in the LTE system and backward compatibility with the LTE system, a band for the LTE-A system is divided into “unit bands” of 20 MHz or less, which is a support bandwidth for the LTE system. That is, the “unit band” herein is a band having a width of maximum 20 MHz and defined as a base unit of a communication band. Furthermore, a “unit band” in a downlink (hereinafter, referred to as “downlink unit band”) may be defined as a band divided by downlink frequency band information included in the BCH broadcasted from the base station, or a band defined by a dispersive width when the downlink control channel (PDCCH) is dispersed and arranged in the frequency domain. Further, a “unit band” in an uplink (hereinafter, referred to as “uplink unit band”) may be defined as a band divided by uplink frequency band information included in the BCH broadcasted from the base station, or as a base unit of a communication band of 20 MHz or less, which includes a physical uplink shared channel (PUSCH) region near the center thereof and PUCCHs for the LTE at both ends thereof. Furthermore, in 3GPP LTE-Advanced, the “unit band” may also be expressed as “component carrier(s)” in English.
The LTE-A system supports communication using a band that bundles several unit bands, so-called “carrier aggregation.” Since throughput requirements for an uplink are generally different from throughput requirements for a downlink, in the LTE-A system, carrier aggregation in which the number of unit bands set for a terminal supporting arbitrary LTE-A system (hereinafter referred to as “LTE-A terminal”) is different between the uplink and the downlink, so-called “asymmetric carrier aggregation” is being discussed. Cases are also supported where the number of unit bands is asymmetric between the uplink and the downlink, and different unit bands have different frequency bandwidths.
FIGS. 3A and 3B are diagrams illustrating asymmetric carrier aggregation applied to individual terminals and a control sequence thereof. FIGS. 3A and 3B illustrates an example in which a bandwidth and the number of unit bands are symmetric between an uplink and a downlink in a base station.
In FIG. 3B, a setting (configuration) is made for terminal 1 such that carrier aggregation is performed using two downlink unit bands and one uplink unit band on the left side, whereas a setting is made for terminal 2 such that although the two same downlink unit bands as those in terminal 1 are used, the uplink unit band on the right side is used for uplink communication.
Focusing attention on terminal 1, signals are transmitted/received between an LTE-A base station and an LTE-A terminal configuring an LTE-A system according to a sequence diagram illustrated in FIG. 3B. As illustrated in FIG. 3A, (1) terminal 1 is synchronized with the downlink unit band (DL CC1) on the left side illustrated in FIG. 3B when communication with the base station starts, and reads information of the uplink unit band which forms a pair with the downlink unit band on the left side from a broadcast signal called “system information block type 2 (SIB2).” (2) Using this uplink unit band (UL CC1), terminal 1 starts communication with the base station by transmitting, for example, a connection request to the base station. (3) Upon deciding that a plurality of downlink unit bands need to be assigned to the terminal, the base station instructs the terminal to add a downlink unit band (DL CC2). In this case, however, the number of uplink unit bands does not increase, and terminal 1 which is an individual terminal starts asymmetric carrier aggregation.
Furthermore, in the LTE-A to which the carrier aggregation is applied, a terminal may receive a plurality of downlink data in a plurality of downlink unit bands at a time. In the LTE-A, a channel selection (which may be also referred to as “multiplexing” or “code selection”) is being discussed as one of methods of transmitting a plurality of response signals in response to the plurality of downlink data. In the channel selection, not only a symbol used for a response signal but also a resource to which the response signal is mapped are changed according to a pattern of an error detection result on the plurality of downlink data. That is, the channel selection is a technique that changes not only a phase point (that is, a constellation point) of the response signal but also a resource used for transmitting the response signal based on whether each of response signals in response to a plurality of downlink data received in a plurality of downlink unit bands is ACK or NACK as illustrated in FIG. 4 (see Non-Patent Literatures 5, 6, and 7).
Here, ARQ control based on the channel selection when the above-described asymmetric carrier aggregation is applied to a terminal will be described below with reference to FIG. 4.
For example, as illustrated in FIG. 4, when a unit band group (which may be expressed as “component carrier set” in English) configured with downlink unit bands 1 and 2 and uplink unit band 1 is set for terminal 1, downlink resource assignment information is transmitted from the base station to terminal 1 through respective PDCCHs of downlink unit bands 1 and 2, and then downlink data is transmitted using a resource corresponding to the downlink resource assignment information.
When the terminal succeeds in receiving the downlink data at unit band 1 and fails to receive the downlink data at unit band 2 (that is, when a response signal of unit band 1 is ACK and a response signal of unit band 2 is NACK), the response signal is mapped to a PUCCH resource included in PUCCH region 1, and a first phase point (e.g. a phase point (1, 0)) is used as a phase point of the response signal. Further, when the terminal succeeds in receiving the downlink data at unit band 1 and also succeeds in receiving the downlink data at unit band 2, the response signal is mapped to a PUCCH resource included in PUCCH region 2, and the first phase point is used. That is, when there are two downlink unit bands, there are four error detection result patterns, so that the four patterns can be represented by combinations of two resources and two types of phase points.